US20050259869A1 - Method for analyzing the color deviations in images using an image sensor - Google Patents
Method for analyzing the color deviations in images using an image sensor Download PDFInfo
- Publication number
- US20050259869A1 US20050259869A1 US10/521,762 US52176205A US2005259869A1 US 20050259869 A1 US20050259869 A1 US 20050259869A1 US 52176205 A US52176205 A US 52176205A US 2005259869 A1 US2005259869 A1 US 2005259869A1
- Authority
- US
- United States
- Prior art keywords
- image sensor
- color channel
- color
- reference data
- sensor signal
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims description 41
- 238000004364 calculation method Methods 0.000 claims description 22
- 238000007689 inspection Methods 0.000 claims description 22
- 230000003595 spectral effect Effects 0.000 claims description 16
- 230000035945 sensitivity Effects 0.000 claims description 10
- 230000009466 transformation Effects 0.000 claims description 9
- 239000003086 colorant Substances 0.000 claims description 8
- 230000004456 color vision Effects 0.000 description 6
- 230000008447 perception Effects 0.000 description 5
- 210000001525 retina Anatomy 0.000 description 5
- 230000005284 excitation Effects 0.000 description 4
- 230000004438 eyesight Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000008485 antagonism Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 3
- 210000002569 neuron Anatomy 0.000 description 3
- 238000010521 absorption reaction Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 230000007704 transition Effects 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000003703 image analysis method Methods 0.000 description 1
- 210000000697 sensory organ Anatomy 0.000 description 1
- 230000001960 triggered effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J3/00—Spectrometry; Spectrophotometry; Monochromators; Measuring colours
- G01J3/46—Measurement of colour; Colour measuring devices, e.g. colorimeters
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J3/00—Spectrometry; Spectrophotometry; Monochromators; Measuring colours
- G01J3/46—Measurement of colour; Colour measuring devices, e.g. colorimeters
- G01J3/462—Computing operations in or between colour spaces; Colour management systems
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J3/00—Spectrometry; Spectrophotometry; Monochromators; Measuring colours
- G01J3/46—Measurement of colour; Colour measuring devices, e.g. colorimeters
- G01J3/465—Measurement of colour; Colour measuring devices, e.g. colorimeters taking into account the colour perception of the eye; using tristimulus detection
Definitions
- the present invention is directed to methods for using an image sensor for analyzing color deviations in images .
- the image signal received from the image sensor is analyzed pixel by pixel.
- the trichromatic model that is used most frequently in technology for describing additive color images is the RGB model.
- RGB model By utilization of the RGB model, an image space is described by the three basic or primary colors, red, green and blue. It is a particular disadvantage, in connection with this model, that the color description performed by the RGB model does not correspond to the color perception of the human eye, since the reaction of the human perception, i.e. the detection through the sensory organs, is not taken into consideration in the RGB model.
- RGB signals are non-linearly transformed into color signal values L, C1, C2.
- a method for classifying color images by the use of fuzzy logic is known from DE 198 38 806 A1.
- USSN 2002/021,444 A1 describes calculation specifications for image processing with the use of weighting factors for the individual color components.
- the object of the present invention is directed to providing methods using an image sensor for analyzing color deviations of images.
- this object is attained by providing an image sensor that is used to analyze color deviations of images.
- An image signal received from the image sensor is analyzed pixel by pixel.
- An image sensor signal is generated for each of the several color channels.
- Image sensor signals from two of the channels are linked by a calculation specification and an output signal of a first compensation channel is generated.
- Linkage of the image sensor channel of the third color channel with the image sensor signals of the first and second color channels is accomplished by using a second calculation specification to generate an output signal of a second compensation color channel.
- the maximum absorption of the S-cone type lies in the blue range, namely at 420 nm.
- the M-cone type absorbs maximally in the green range, namely at 534 nm.
- the L-cone type has its absorption maximum at 564 nm in the yellow/red spectral range.
- Color antagonism means that defined colors can never be seen in transitions, so that no color transition between these colors is possible. Colors which show color antagonism are called compensation or complementary colors. The color pairs red/green and blue/yellow, as well as black/white should be mentioned here.
- the different spectral distribution of the light which for example is a function of the weather or daylight conditions, is compensated.
- the compensation color model assumes that the cones are arranged in receptive fields, namely in blue/yellow fields and in red/green fields.
- Receptive fields are understood here to be neurons, and describe the manner in which the light signals coming from the cones are further processed by the neurons.
- Substantially two types of receptive fields are responsible for color vision. The first receptive field obtains its input from the L- and M-cones. The second receptive field obtains its input from the S-cones, together with differently weighted signals from the L- and M-cones. It is assumed that subtractive color mixing of the excitations of the cones is performed in the plane of the neurons or in these receptive fields.
- an image signal that is received from the image sensor, is analyzed pixel by pixel in a manner which is generally known .
- the image signal is recorded by the image sensor in three color channels, which are separated from each other, and which are used for reproducing the three cone types in the human eye with their different spectral sensitivities.
- Each one of the three color channels has a characteristic spectral sensitivity.
- the two receptive fields, representing the second stage of color processing in human vision are simulated by an appropriate linkage of the image sensor signals of the three color channels, which are separated from each other.
- the red/green field of human color perception represents the first compensation color channel.
- the output signal of the first compensation color channel is generated by the linkage of the image sensor signal of a first color signal with the image sensor signal of a second color signal.
- Linkage is performed by the utilization of a calculation specification consisting of at least one arithmetic rule.
- the blue/yellow field is generated by linking the image sensor signal of a third color channel with a combination of the image sensor signals of the first and second color channels.
- the blue/yellow field corresponds to the second compensation color channel.
- the output signal of the second compensation color channel is generated by use of the above described linkage.
- Linkage is performed by a second calculation specification consisting of at least one arithmetic rule.
- a classification of the output signals from the two compensation color channels takes place in the next step for evaluating the examined pixel. It is decided by this classification whether the image content of the examined pixel corresponds to a defined class, by the use of which classification an acceptable/unacceptable determination can be made.
- a preferred embodiment of the present invention is one in which the three color channels correspond to the basic colors of the RGB model, namely red, green and blue. This has the advantage that it is possible to use a widely distributed color model.
- the spectral sensitivity of every color channel can be matched to spectral sensitivities of the cones.
- One possibility consists in that one arithmetic rule of the first calculation specification provides the formation of a weighted differentiation between the image sensor signals of the first color channel, and/or an arithmetic rule of the second calculation specification provides the formation of a weighted differentiation between the weighted sum of the image sensor signals from the first and second image sensor channels and the third color channel.
- At least one signal in at least one compensation color channel is subjected to a transformation specification, and in particular is subjected to a non-linear transformation specification, after and/or prior to the linkage.
- a transformation has the particular advantage that the digital character of electronically generated recordings can be taken into consideration.
- the transformation specifications it is also possible to transform a signal from the color space into a space in which the excitation of the cones can be described.
- the signals in both of the compensation color channels are subjected to a transformation.
- the receptive fields are characterized by a low pass behavior. It is thus useful if at least one signal in at least one compensation color channel is filtered by the use of a low pass filter.
- the output signal of each compensation color channel is filtered by the use of a low pass filter.
- the method for analyzing color deviations in images has a learning mode of operation and an inspection mode of operation.
- the learning mode of operation at least one reference image is analyzed pixel by pixel.
- the two output signals from the two compensation color channels formed by the reference image are stored in a reference memory. This means that the image content of the reference image is actually recorded in three color channels, that the image signals of each color channel are matched in accordance with perception, and that these image signals are subsequently linked with each other, in accordance with the compensation color model.
- the output signals from each compensation color channel are stored, pixel by pixel, in the reference data memory.
- the output signals from the appropriate pixel, that are generated by a test image are compared with the corresponding value in the reference data memory, and a classification decision is made regarding the acceptability of the test image.
- the values stored in the reference data memory are formed by analyzing several reference data sets, so that, for each value in the reference data memory, a permissible tolerance window is provided, and within which tolerance window an output signal value of a compensation color channel, generated during the image inspection, can fluctuate.
- the reference variable of the output signal of a compensation color channel can be determined, for example, by forming the average arithmetic value of the individual values resulting from the reference data sets.
- the tolerance window can be determined, for example, from the minimum and maximum values of the output signal, or from the standard deviation of the output signals of the compensation color channels generated from the examined reference images.
- FIG. 1 a schematic representation of a method for analyzing color deviations in print images with a compensation color model in accordance with the present invention
- FIG. 2 a flow chart of the learning and inspection mode.
- recording of image signals of an image is accomplished by an image sensor and takes place in three color channels 01 , 02 , 03 , which are separated from each other.
- the three separated color channels 01 , 02 , 03 are the color channels red 01 , green 02 and blue 03 .
- Each one of the three color channels has an adjustable spectral sensitivity. This has, as an advantage, that the characteristic of each color channel 01 , 02 , 03 recorded by the image sensor can be matched to the conditions of the existing set of problems. Thus, it is possible, for example, to match the spectral sensitivity of a color channel 01 , 02 , 03 to the spectral sensitivity of the appropriate cone of the retina of the human eye.
- the spectral content of an image is analyzed pixel by pixel.
- the image sensor signals of the color channels 01 , 02 , 03 are linked with each other in accordance with the method of the present invention.
- each image sensor signal in two resultant compensation color channel 07 , 08 is subjected to a non-linear transformation 09 .
- the digital character of the electronically generated recordings is taken into consideration by this.
- the image sensor signals of the red color channel 01 and the image sensor signals of the green color channel 02 are linked with each other by use of the first calculation specification 04 .
- the image sensor signal of the blue color channel 03 is linked with the minimum one signal 14 of the image sensor signals of the red color channel 01 and of the green color channel 02 by use of the calculation specification 06 .
- the receptive fields of the human retina are characterized by a low pass behavior. Accordingly, in the present preferred embodiment, the signals obtained by each linkage are subjected to a low pass filtering process 16 using a Gauss low pass filter.
- FIG. 2 is a schematic depiction of the actual inspection of printed products, which actual inspection is performed in two stages, namely in a first, learning mode 17 and a downstream connected second, inspection mode 18 .
- the aim of the inspection of printed products, in the learning mode 17 is the generation, pixel by pixel, of reference data values 19 , which are then compared, in the subsequent inspection mode 18 , with the output signals 12 , 13 of the compensation color channels 07 , 08 of the appropriate inspected pixels.
- the image contents of one reference image 21 or the image contents of several reference images 21 , are analyzed.
- the image contents of each pixel in the reference image or images are entered into three color channels 01 , 02 , 03 and a subsequent adaptation with respect to perception of the image signals of each color channel 01 , 02 , 03 , is performed. Further processing of the image sensor signals, in accordance with the above-described compensation color method, is performed. Then the output signals 12 , 13 of the compensation color channels 07 , 08 obtained for each pixel of the reference are stored in a reference data memory. It is useful, for the purpose of also taking into consideration permissible fluctuations of the reference images 21 , if several reference images 21 are considered in the learning mode 17 .
- the reference data values 19′, 19′′ of each pixel that is stored in the reference data memory can have a defined permissible fluctuation tolerance.
- the fluctuation tolerance can be determined either by the use of minimum/maximum values, or from the standard deviation of the obtained data of the image contents of the reference images 21 of each pixel.
- a comparison, on a pixel by pixel basis, of the output signals 12 , 13 of the first and second compensation color channels 07 , 08 of an inspection image 22 , with the reference data values 19′, 19′′ taken from the reference data memory takes place.
- the comparison of the image values with the reference values can be performed by the use of a linear or a non-linear classifier 23 , and in particular by the use of threshold value classifiers, Euclidic distance classifiers, Bayes classifiers, fuzzy classifiers, or artificial neuronic networks. Subsequently an acceptable/unacceptable decision is made regarding the inspection image 22 .
Abstract
Description
- This U.S. patent application is the U.S. national phase, under 35 USC 371 of PCT/DE2003/002146, filed Jun. 27, 2003, published as WO 2004/017034 A1 on Feb. 26, 2004 and claiming priority to DE 102 34 085.4 filed Jul. 26, 2002, the disclosures of which are expressly incorporated herein by reference.
- The present invention is directed to methods for using an image sensor for analyzing color deviations in images . The image signal received from the image sensor is analyzed pixel by pixel.
- The trichromatic model that is used most frequently in technology for describing additive color images is the RGB model. By utilization of the RGB model, an image space is described by the three basic or primary colors, red, green and blue. It is a particular disadvantage, in connection with this model, that the color description performed by the RGB model does not correspond to the color perception of the human eye, since the reaction of the human perception, i.e. the detection through the sensory organs, is not taken into consideration in the RGB model.
- DE 44 19 395 A1 discloses a method for analyzing color images by the use of an image sensor, whose image signals are analyzed pixel by pixel. In the process, the image signals are separated in accordance with colorfulness and with brightness.
- DE 692 24 812 T2 describes a method for image processing. RGB signals are non-linearly transformed into color signal values L, C1, C2.
- A method for classifying color images by the use of fuzzy logic is known from DE 198 38 806 A1.
- USSN 2002/021,444 A1 describes calculation specifications for image processing with the use of weighting factors for the individual color components.
- A method for analyzing color deviations in images is set forth in the article by Michael J. Swain and Dana H. Ballard, “Color Indexing,” Internat'l. Journal of Computer Vision, 7:1, 11-32 (1991), XP000248428. Compensation color channels are generated from color channels, which compensation color channels correspond to the receptive fields of the human eye. These compensation color channels are used for preparing histograms.
- EP 0 473 432 B1 discloses an image analysis method which provides a transformation with several input signals with associated calculation specifications.
- The object of the present invention is directed to providing methods using an image sensor for analyzing color deviations of images.
- In accordance with the present invention, this object is attained by providing an image sensor that is used to analyze color deviations of images. An image signal received from the image sensor is analyzed pixel by pixel. An image sensor signal is generated for each of the several color channels. Image sensor signals from two of the channels are linked by a calculation specification and an output signal of a first compensation channel is generated. Linkage of the image sensor channel of the third color channel with the image sensor signals of the first and second color channels is accomplished by using a second calculation specification to generate an output signal of a second compensation color channel. These color channels are classified.
- Three cone types, which each absorb light in different spectral ranges, exist in the human eye. The maximum absorption of the S-cone type lies in the blue range, namely at 420 nm. The M-cone type absorbs maximally in the green range, namely at 534 nm. The L-cone type has its absorption maximum at 564 nm in the yellow/red spectral range. Seeing by use of three types of cones is called trichromatic vision. An individual color perception is triggered by the different excitation strengths of the individual cone types. Identical excitation of all three types of cones leads to the perception of the color white. However, by use of the trichromatic vision model, it is not possible to explain color perception phenomena, such as color antagonism and color constancy, for example.
- Color antagonism means that defined colors can never be seen in transitions, so that no color transition between these colors is possible. Colors which show color antagonism are called compensation or complementary colors. The color pairs red/green and blue/yellow, as well as black/white should be mentioned here.
- With color constancy, the different spectral distribution of the light, which for example is a function of the weather or daylight conditions, is compensated.
- Hering developed his compensation color theory in 1920 to explain these color perception phenomena in a way that was different from the classic trichromatic color model. The compensation color model assumes that the cones are arranged in receptive fields, namely in blue/yellow fields and in red/green fields. Receptive fields are understood here to be neurons, and describe the manner in which the light signals coming from the cones are further processed by the neurons. Substantially two types of receptive fields are responsible for color vision. The first receptive field obtains its input from the L- and M-cones. The second receptive field obtains its input from the S-cones, together with differently weighted signals from the L- and M-cones. It is assumed that subtractive color mixing of the excitations of the cones is performed in the plane of the neurons or in these receptive fields.
- In the process for analyzing color deviations of printed images, an image signal, that is received from the image sensor, is analyzed pixel by pixel in a manner which is generally known . In accordance with the present invention, the image signal is recorded by the image sensor in three color channels, which are separated from each other, and which are used for reproducing the three cone types in the human eye with their different spectral sensitivities. Each one of the three color channels has a characteristic spectral sensitivity. The two receptive fields, representing the second stage of color processing in human vision, are simulated by an appropriate linkage of the image sensor signals of the three color channels, which are separated from each other. In the technical model, the red/green field of human color perception represents the first compensation color channel. The output signal of the first compensation color channel is generated by the linkage of the image sensor signal of a first color signal with the image sensor signal of a second color signal. Linkage is performed by the utilization of a calculation specification consisting of at least one arithmetic rule. In the technical model, the blue/yellow field is generated by linking the image sensor signal of a third color channel with a combination of the image sensor signals of the first and second color channels. In the technical model, the blue/yellow field corresponds to the second compensation color channel. The output signal of the second compensation color channel is generated by use of the above described linkage. Linkage is performed by a second calculation specification consisting of at least one arithmetic rule. A classification of the output signals from the two compensation color channels takes place in the next step for evaluating the examined pixel. It is decided by this classification whether the image content of the examined pixel corresponds to a defined class, by the use of which classification an acceptable/unacceptable determination can be made.
- It is without great importance, for the principle of the present invention, in which spectral range the three color channels of the method are located, so long as these three color channels are color channels which are separated from each other. A preferred embodiment of the present invention is one in which the three color channels correspond to the basic colors of the RGB model, namely red, green and blue. This has the advantage that it is possible to use a widely distributed color model.
- To match the spectral sensitivity of each color channel to the spectral perception of the corresponding cone of the retina of the human eye, it is beneficial if the spectral sensitivity of every color channel can be matched to spectral sensitivities of the cones.
- It is of secondary importance, for the principle of the present invention, in which way the two output signals of the compensation color channels are generated. One possibility consists in that one arithmetic rule of the first calculation specification provides the formation of a weighted differentiation between the image sensor signals of the first color channel, and/or an arithmetic rule of the second calculation specification provides the formation of a weighted differentiation between the weighted sum of the image sensor signals from the first and second image sensor channels and the third color channel.
- In accordance with a further preferred embodiment of the present invention, at least one signal in at least one compensation color channel is subjected to a transformation specification, and in particular is subjected to a non-linear transformation specification, after and/or prior to the linkage. A transformation has the particular advantage that the digital character of electronically generated recordings can be taken into consideration. By the use of the transformation specifications, it is also possible to transform a signal from the color space into a space in which the excitation of the cones can be described. In many preferred embodiments, the signals in both of the compensation color channels are subjected to a transformation.
- With human vision, the receptive fields are characterized by a low pass behavior. It is thus useful if at least one signal in at least one compensation color channel is filtered by the use of a low pass filter. In accordance with a particularly preferred embodiment of the present invention, the output signal of each compensation color channel is filtered by the use of a low pass filter.
- In accordance with a particularly preferred embodiment of the present invention, the method for analyzing color deviations in images, using an image sensor, has a learning mode of operation and an inspection mode of operation. In the course of the learning mode of operation, at least one reference image is analyzed pixel by pixel. The two output signals from the two compensation color channels formed by the reference image are stored in a reference memory. This means that the image content of the reference image is actually recorded in three color channels, that the image signals of each color channel are matched in accordance with perception, and that these image signals are subsequently linked with each other, in accordance with the compensation color model. Then, the output signals from each compensation color channel are stored, pixel by pixel, in the reference data memory. In a subsequent inspection mode, the output signals from the appropriate pixel, that are generated by a test image, are compared with the corresponding value in the reference data memory, and a classification decision is made regarding the acceptability of the test image.
- In order to take permissible fluctuations of the image content, as well as to take fluctuations of the conditions existing during the image recording, into consideration, it is useful if the values stored in the reference data memory are formed by analyzing several reference data sets, so that, for each value in the reference data memory, a permissible tolerance window is provided, and within which tolerance window an output signal value of a compensation color channel, generated during the image inspection, can fluctuate. In this case, the reference variable of the output signal of a compensation color channel can be determined, for example, by forming the average arithmetic value of the individual values resulting from the reference data sets. The tolerance window can be determined, for example, from the minimum and maximum values of the output signal, or from the standard deviation of the output signals of the compensation color channels generated from the examined reference images.
- Preferred embodiments of the present invention are represented in the drawings and will be described in greater detail in what follows.
- Shown are in:
-
FIG. 1 , a schematic representation of a method for analyzing color deviations in print images with a compensation color model in accordance with the present invention, and in -
FIG. 2 , a flow chart of the learning and inspection mode. - As may be seen in
FIG. 1 , recording of image signals of an image is accomplished by an image sensor and takes place in threecolor channels color channels color channel color channel - In the method for analyzing the color deviations in images in accordance with the present invention, the spectral content of an image is analyzed pixel by pixel. For modeling the two receptive fields, red/green and blue/yellow of the human eye, the image sensor signals of the
color channels calculation specifications compensation color channel non-linear transformation 09. The digital character of the electronically generated recordings is taken into consideration by this. Subsequently, each signal is weighted with a coefficient KI (I=1 . . . 4). By the use of this process, it is achieved that a pure intensity change of the initial image does not provide a contribution to one of the output signals 12, 13 of the first and secondcompensation color channels compensation color channels calculation specifications color channels output signal 12 of the red/greencompensation color channel 07, the image sensor signals of thered color channel 01 and the image sensor signals of thegreen color channel 02 are linked with each other by use of thefirst calculation specification 04. To form theoutput signal 13 of the blue/yellowcompensation color channel 08, in the present preferred embodiment the image sensor signal of theblue color channel 03 is linked with the minimum onesignal 14 of the image sensor signals of thered color channel 01 and of thegreen color channel 02 by use of thecalculation specification 06. The receptive fields of the human retina are characterized by a low pass behavior. Accordingly, in the present preferred embodiment, the signals obtained by each linkage are subjected to a lowpass filtering process 16 using a Gauss low pass filter. -
FIG. 2 is a schematic depiction of the actual inspection of printed products, which actual inspection is performed in two stages, namely in a first, learning mode 17 and a downstream connected second, inspection mode 18. The aim of the inspection of printed products, in the learning mode 17, is the generation, pixel by pixel, of reference data values 19, which are then compared, in the subsequent inspection mode 18, with the output signals 12, 13 of thecompensation color channels reference image 21, or the image contents ofseveral reference images 21, are analyzed. The image contents of each pixel in the reference image or images are entered into threecolor channels color channel compensation color channels reference images 21, ifseveral reference images 21 are considered in the learning mode 17. Because of this consideration of several reference images, it is thus possible for the reference data values 19′, 19″ of each pixel that is stored in the reference data memory to have a defined permissible fluctuation tolerance. The fluctuation tolerance can be determined either by the use of minimum/maximum values, or from the standard deviation of the obtained data of the image contents of thereference images 21 of each pixel. - Then, in the actual inspection mode 18, a comparison, on a pixel by pixel basis, of the output signals 12, 13 of the first and second
compensation color channels inspection image 22, with the reference data values 19′, 19″ taken from the reference data memory takes place. The comparison of the image values with the reference values can be performed by the use of a linear or anon-linear classifier 23, and in particular by the use of threshold value classifiers, Euclidic distance classifiers, Bayes classifiers, fuzzy classifiers, or artificial neuronic networks. Subsequently an acceptable/unacceptable decision is made regarding theinspection image 22. - While a preferred embodiment of a method for analyzing the color deviations in images using an image sensor, in accordance with the present invention, has been set forth fully and completely hereinabove, it will be apparent to one of skill in the art that various changes in, for example, the specific image being inspected, the procedures to be followed once the image has been inspected, and the like could be made without departing from the true spirit and scope of the present invention which is accordingly to be limited only by the following claims.
Claims (23)
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
DE10234085.4 | 2002-07-26 | ||
DE10234085A DE10234085B4 (en) | 2002-07-26 | 2002-07-26 | Method for analyzing color deviations of images with an image sensor |
PCT/DE2003/002146 WO2004017034A1 (en) | 2002-07-26 | 2003-06-27 | Method for analyzing the color deviations in images using an image sensor |
Publications (2)
Publication Number | Publication Date |
---|---|
US20050259869A1 true US20050259869A1 (en) | 2005-11-24 |
US7609878B2 US7609878B2 (en) | 2009-10-27 |
Family
ID=30469124
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US10/521,762 Expired - Fee Related US7609878B2 (en) | 2002-07-26 | 2003-06-27 | Method for analyzing the color deviations in images using an image sensor |
Country Status (7)
Country | Link |
---|---|
US (1) | US7609878B2 (en) |
EP (1) | EP1525442B1 (en) |
AT (1) | ATE511632T1 (en) |
AU (1) | AU2003250762A1 (en) |
DE (1) | DE10234085B4 (en) |
ES (1) | ES2364650T3 (en) |
WO (1) | WO2004017034A1 (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070223789A1 (en) * | 2004-04-23 | 2007-09-27 | Carsten Diederichs | Method for Evaluating the Quality of a Printed Matter, Provided by a Printing Machine |
US20080295724A1 (en) * | 2005-11-25 | 2008-12-04 | Volker Lohweg | Method For Detection Of Occurrence Of Printing Errors On Printed Substrates During Processing Thereof On A Printing Press |
US20150317787A1 (en) * | 2014-03-28 | 2015-11-05 | Intelliview Technologies Inc. | Leak detection |
CN110667258A (en) * | 2018-07-03 | 2020-01-10 | 海德堡印刷机械股份公司 | Print quality analysis by means of neural networks |
US10902644B2 (en) * | 2017-08-07 | 2021-01-26 | Samsung Display Co., Ltd. | Measures for image testing |
US11425281B1 (en) * | 2016-09-07 | 2022-08-23 | Google Llc | Color image processing using models of overlapped response spectra for retinal cone cells |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE102004021047B3 (en) * | 2004-04-29 | 2005-10-06 | Koenig & Bauer Ag | Method for comparing an image with at least one reference image |
EP1790473A1 (en) | 2005-11-25 | 2007-05-30 | Kba-Giori S.A. | Method for detection of occurrence of printing errors on printed substrates during processing thereof on a printing press |
US10600213B2 (en) * | 2016-02-27 | 2020-03-24 | Focal Sharp, Inc. | Method and apparatus for color-preserving spectrum reshape |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5237517A (en) * | 1990-08-29 | 1993-08-17 | Xerox Corporation | Color printing having a highlight color image mapped from a full color image |
US5268753A (en) * | 1990-08-20 | 1993-12-07 | Brother Kogyo Kabushiki Kaisha | Color image forming apparatus and method thereof with color-correcting operation |
US5710644A (en) * | 1992-01-06 | 1998-01-20 | Canon Kabushiki Kaisha | Color expressing method and image processing apparatus thereof |
US5999636A (en) * | 1997-10-10 | 1999-12-07 | Printprobe Technology, Llc | Apparatus and process for inspecting print material |
US20020021444A1 (en) * | 1997-12-18 | 2002-02-21 | Chromatics Colorsciences Inc., | Color measurement system with color index for skin, teeth, hair and material substances |
US6486981B1 (en) * | 1993-07-27 | 2002-11-26 | Canon Kabushiki Kaisha | Color image processing method and apparatus thereof |
US6751348B2 (en) * | 2001-03-29 | 2004-06-15 | Fotonation Holdings, Llc | Automated detection of pornographic images |
US6911963B2 (en) * | 2000-12-21 | 2005-06-28 | Kabushiki Kaisha Toshiba | Field-sequential color display unit and display method |
US6950554B2 (en) * | 2000-07-18 | 2005-09-27 | Olympus Optical Co., Ltd. | Learning type image classification apparatus, method thereof and processing recording medium on which processing program is recorded |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR940008510A (en) * | 1992-09-14 | 1994-04-29 | 김광호 | Video Signal Processing Circuit of Solid State Imaging Device (CCD) Color Video Camera |
DE4419395C2 (en) * | 1994-05-30 | 2000-09-28 | Graphikon Gmbh | Method and device for analyzing and processing color images |
DE19838806A1 (en) * | 1998-08-26 | 2000-03-02 | Max Planck Gesellschaft | Method and device for detecting object colors |
-
2002
- 2002-07-26 DE DE10234085A patent/DE10234085B4/en not_active Revoked
-
2003
- 2003-06-27 ES ES03787613T patent/ES2364650T3/en not_active Expired - Lifetime
- 2003-06-27 AU AU2003250762A patent/AU2003250762A1/en not_active Abandoned
- 2003-06-27 EP EP03787613A patent/EP1525442B1/en not_active Revoked
- 2003-06-27 WO PCT/DE2003/002146 patent/WO2004017034A1/en not_active Application Discontinuation
- 2003-06-27 US US10/521,762 patent/US7609878B2/en not_active Expired - Fee Related
- 2003-06-27 AT AT03787613T patent/ATE511632T1/en active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5268753A (en) * | 1990-08-20 | 1993-12-07 | Brother Kogyo Kabushiki Kaisha | Color image forming apparatus and method thereof with color-correcting operation |
US5237517A (en) * | 1990-08-29 | 1993-08-17 | Xerox Corporation | Color printing having a highlight color image mapped from a full color image |
US5710644A (en) * | 1992-01-06 | 1998-01-20 | Canon Kabushiki Kaisha | Color expressing method and image processing apparatus thereof |
US6486981B1 (en) * | 1993-07-27 | 2002-11-26 | Canon Kabushiki Kaisha | Color image processing method and apparatus thereof |
US5999636A (en) * | 1997-10-10 | 1999-12-07 | Printprobe Technology, Llc | Apparatus and process for inspecting print material |
US20020021444A1 (en) * | 1997-12-18 | 2002-02-21 | Chromatics Colorsciences Inc., | Color measurement system with color index for skin, teeth, hair and material substances |
US6950554B2 (en) * | 2000-07-18 | 2005-09-27 | Olympus Optical Co., Ltd. | Learning type image classification apparatus, method thereof and processing recording medium on which processing program is recorded |
US6911963B2 (en) * | 2000-12-21 | 2005-06-28 | Kabushiki Kaisha Toshiba | Field-sequential color display unit and display method |
US6751348B2 (en) * | 2001-03-29 | 2004-06-15 | Fotonation Holdings, Llc | Automated detection of pornographic images |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070223789A1 (en) * | 2004-04-23 | 2007-09-27 | Carsten Diederichs | Method for Evaluating the Quality of a Printed Matter, Provided by a Printing Machine |
US7672486B2 (en) | 2004-04-23 | 2010-03-02 | Koenig & Bauer Aktiengesellschaft | Method for evaluating the quality of a printed matter, provided by a printing machine |
US20080295724A1 (en) * | 2005-11-25 | 2008-12-04 | Volker Lohweg | Method For Detection Of Occurrence Of Printing Errors On Printed Substrates During Processing Thereof On A Printing Press |
US8613254B2 (en) | 2005-11-25 | 2013-12-24 | Kba-Notasys Sa | Method for detection of occurrence of printing errors on printed substrates during processing thereof on a printing press |
US20150317787A1 (en) * | 2014-03-28 | 2015-11-05 | Intelliview Technologies Inc. | Leak detection |
US10234354B2 (en) * | 2014-03-28 | 2019-03-19 | Intelliview Technologies Inc. | Leak detection |
US11425281B1 (en) * | 2016-09-07 | 2022-08-23 | Google Llc | Color image processing using models of overlapped response spectra for retinal cone cells |
US10902644B2 (en) * | 2017-08-07 | 2021-01-26 | Samsung Display Co., Ltd. | Measures for image testing |
CN110667258A (en) * | 2018-07-03 | 2020-01-10 | 海德堡印刷机械股份公司 | Print quality analysis by means of neural networks |
Also Published As
Publication number | Publication date |
---|---|
EP1525442A1 (en) | 2005-04-27 |
US7609878B2 (en) | 2009-10-27 |
DE10234085A1 (en) | 2004-02-19 |
DE10234085B4 (en) | 2012-10-18 |
AU2003250762A1 (en) | 2004-03-03 |
ES2364650T3 (en) | 2011-09-08 |
ATE511632T1 (en) | 2011-06-15 |
EP1525442B1 (en) | 2011-06-01 |
WO2004017034A1 (en) | 2004-02-26 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7447353B2 (en) | Methods for qualitative evaluation of a material with at least one identification characteristic | |
JP3417051B2 (en) | Color information processing method and apparatus using feature parameter values independent of light source | |
US7609878B2 (en) | Method for analyzing the color deviations in images using an image sensor | |
US5559604A (en) | Color matching apparatus for reproducing the same color under different illuminants | |
CN106651966B (en) | Picture color identification method and system | |
CN105574514A (en) | Greenhouse immature tomato automatic identification method | |
Milić et al. | A content-dependent naturalness-preserving daltonization method for dichromatic and anomalous trichromatic color vision deficiencies | |
CN103310201A (en) | Identification method for mixed colors of targets | |
Wang et al. | Facial image medical analysis system using quantitative chromatic feature | |
CN107135384A (en) | White balance adjusting method, device, image processing terminal and storage medium | |
US20070172118A1 (en) | Method for reducing enhancement of artifacts and noise in image color enhancement | |
CN114119551A (en) | Quantitative analysis method for human face image quality | |
Mohammed et al. | Image segmentation for skin detection | |
Lin | An automatic white balance method based on edge detection | |
CN110891172B (en) | Dead pixel detection and correction device | |
Istiadi et al. | Identification of tempe fermentation maturity using principal component analysis and k-nearest neighbor | |
CN100458423C (en) | Detecting method for color printing quality | |
Titova et al. | Food quality evaluation according to their color characteristics | |
CN115292722A (en) | Model safety detection method and device based on different color spaces | |
CN114926661A (en) | Textile surface color data processing and identifying method and system | |
CN115049530A (en) | Method, apparatus and system for debugging image signal processor | |
JP3420629B2 (en) | Image evaluation method and color image input device | |
Dehesa‐González et al. | Lighting source classification applied in color images to contrast enhancement | |
Busin et al. | Colour space selection for unsupervised colour image segmentation by analysis of connectedness properties | |
Kucuk et al. | Comparison of Regression Methods and Neural Networks for Colour Corrections |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: KOENIG & BAUER AKTIENGESELLSCHAFT, GERMANY Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:LOHWEG, VOLKER;WILLEKE, HARALD HEINRICH;REEL/FRAME:016801/0010;SIGNING DATES FROM 20050106 TO 20050110 |
|
STCF | Information on status: patent grant |
Free format text: PATENTED CASE |
|
FPAY | Fee payment |
Year of fee payment: 4 |
|
AS | Assignment |
Owner name: KOENIG & BAUER AG, GERMANY Free format text: CHANGE OF NAME;ASSIGNOR:KOENIG & BAUER AKTIENGESELLSCHAFT;REEL/FRAME:036987/0915 Effective date: 20150521 |
|
FPAY | Fee payment |
Year of fee payment: 8 |
|
FEPP | Fee payment procedure |
Free format text: MAINTENANCE FEE REMINDER MAILED (ORIGINAL EVENT CODE: REM.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
LAPS | Lapse for failure to pay maintenance fees |
Free format text: PATENT EXPIRED FOR FAILURE TO PAY MAINTENANCE FEES (ORIGINAL EVENT CODE: EXP.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY |
|
STCH | Information on status: patent discontinuation |
Free format text: PATENT EXPIRED DUE TO NONPAYMENT OF MAINTENANCE FEES UNDER 37 CFR 1.362 |
|
FP | Lapsed due to failure to pay maintenance fee |
Effective date: 20211027 |